Intelligent pavement condition survey:Overview of current researches and practices

Automated pavement condition survey is of critical importance to road network management.There are three primary tasks involved in pavement condition surveys,namely data collection,data processing and condition evaluation.Artificial intelligence(AI)has achieved many breakthroughs in almost every aspect of modern technology over the past decade,and undoubtedly offers a more robust approach to automated pavement condition survey.This article aims to provide a comprehensive review on data collection systems,data processing algorithms and condition evaluation methods proposed between 2010 and 2023 for intelligent pavement condition survey.In particular,the data collection system includes AI-driven hardware devices and automated pavement data collection vehicles.The AI-driven hardware devices including right-of-way(ROW)cameras,ground penetrating radar(GPR)devices,light detection and ranging(LiDAR)devices,and advanced laser imaging systems,etc.These different hardware components can be selectively mounted on a vehicle to simultaneously collect multimedia information about the pavement.In addition,this article pays close attention to the application of artificial intelligence methods in detecting pavement distresses,measuring pavement roughness,identifying pavement rutting,analyzing skid resistance and evaluating structural strength of pavements.Based upon the analysis of a variety of the state-of-the-art artificial intelligence methodologies,remaining challenges and future needs with respect to intelligent pavement condition survey are discussed eventually.

Predicting pavement condition index based on the utilization of machine learning techniques:A case study

Pavement management systems(PMS)are used by transportation government agencies to promote sustainable development and to keep road pavement conditions above the minimum performance levels at a reasonable cost.To accomplish this objective,the pavement condition is monitored to predict deterioration and determine the need for maintenance or rehabilitation at the appropriate time.The pavement condition index(PCI)is a commonly usedmetric to evaluate the pavement's performance.This research aims to create and evaluate prediction models for PCI values using multiple linear regression(MLR),artificial neural networks(ANN),and fuzzy logic inference(FIS)models for flexible pavement sections.The authors collected field data spans for 2018 and 2021.Eight pavement distress factors were considered inputs for predicting PCI values,such as rutting,fatigue cracking,block cracking,longitudinal cracking,transverse cracking,patching,potholes,and delamination.This study evaluates the performance of the three techniques based on the coefficient of determination,root mean squared error(RMSE),and mean absolute error(MAE).The results show that the R2 values of the ANN models increased by 51.32%,2.02%,36.55%,and 3.02%compared toMLR and FIS(2018 and 2021).The error in the PCI values predicted by the ANNmodel was significantly lower than the errors in the prediction by the FIS and MLR models.

Effect of Sample Unit Size on Visually Examining Pavement Condition for Asphalt-Surfaced Roads

Road surface condition evaluation involves the collection of data over pavement surface for different types of distresses. The exercise consumes a lot of resources if the whole road section length is surveyed and may be prone to errors as a result of surveyors＇ fatigue. It is therefore important to develop a representative sample to be used when evaluating road condition manually. This study aimed at determining an adequate sample size for section level as well as a way forward for network level condition evaluation of highways in Nepal. Again the study was conducted to quantify the effects of altering the sample unit size for performing a distress survey according to the PCI （pavement condition index） and SDI （surface distress index） method separately for asphalt surfaced roads. The effect of reducing/increasing sample unit size was investigated adopting visual examination through field survey by eight teams in July, 2015, along the section of Banepa-Bardibas highway. The PCI was then calculated for each sample unit using standard deduct curves and PCI calculation methodology as per SHRP （Strategic Highway Research Program） recommendations and the computation of SDI was done as per DoR （Department of Roads） guidelines. The results show that 13% sample unit are needed for SDI and 21% for PCI computation, however, the results are out of the significant level. This is higher than DoR and SHRP guidelines. Again no strong relationship is observed between SDI and PCI values.

Road pavement condition diagnostics using smartphone-based data crowdsourcing in smart cities

The purpose of the paper is to analyse the effectiveness of a solution known as road condition tool(RCT)based on data crowdsourcing from smartphones users in the transport system.The tool developed by the author of the paper,enabling identification and assessment of road pavement defects by analysing the dynamics of vehicle motion in the road network.Transport system users equipped with a smartphone with the RCT mobile application on board record data of linear accelerations,speed,and vehicle location,and then,without any intervention,send them to the RCT server database in an aggregated form.The aggregated data are processed in the combined time and location criterion,and the road pavement condition assessment index is estimated for fixed 10 m long measuring sections.The measuring sections correspond to the sections of roads defined in the pavement management systems(PMS)used by municipal road infrastructure administration bodies.Both the research in question and the results obtained by the method proposed for purposes of the road pavement condition assessment were compared with a set of reference data of the road infrastructure administration body which conducted surveys using highly specialised measuring equipment.The results of this comparison,performed using binary classifiers,confirm the potential RCT solution proposed by the author.This solution makes it possible to global monitor the road infrastructure condition on a continuous basis via numerous users of the transport system,which guarantees that such an assessment is kept up to date.

Effect of speed hump characteristics on pavement condition

Speed humps are the most common type of traffic calming devices due to their low cost and easy installation. However, in many Egyptian roads, considerable number of these humps is randomly placed without proper engineering studies and justifications. Deteri- oration of pavement condition is observed near these humps. This paper presents a case study applied to collect and analyze visual inspection data for the reason of evaluating the impact of speed humps on pavement condition on intercity rural roads. The paper used 52 speed humps located in an intercity two-lane, two-way road that connects two cities, Tahta and Gerga, in Upper Egypt. The total length of this road is about 34 km. Pavement condition index （PCI）, in road sections, near speed humps in the two directions of travel were calculated from the visual inspection measurements. The characteristics of each speed hump （width, height, and distance from preceding hump） were measured. Using statistical analyses, the correlations between the pavement conditions and hump char- acteristics were examined. Regression analysis models were developed to represent the relationships between pavement conditions and hump characteristics. Generally, the re- sults proved that the pavement conditions are greatly influenced by the presence of speed humps and hump characteristics.

Fast prediction of the mechanical response for layered pavement under instantaneous large impact based on random forest regression

The layered pavements usually exhibit complicated mechanical properties with the effect of complex material properties under external environment.In some cases,such as launching missiles or rockets,layered pavements are required to bear large impulse load.However,traditional methods cannot non-destructively and quickly detect the internal structural of pavements.Thus,accurate and fast prediction of the mechanical properties of layered pavements is of great importance and necessity.In recent years,machine learning has shown great superiority in solving nonlinear problems.In this work,we present a method of predicting the maximum deflection and damage factor of layered pavements under instantaneous large impact based on random forest regression with the deflection basin parameters obtained from falling weight deflection testing.The regression coefficient R^(2)of testing datasets are above 0.94 in the process of predicting the elastic moduli of structural layers and mechanical responses,which indicates that the prediction results have great consistency with finite element simulation results.This paper provides a novel method for fast and accurate prediction of pavement mechanical responses under instantaneous large impact load using partial structural parameters of pavements,and has application potential in non-destructive evaluation of pavement structure.

Design and laboratory evaluation of fog-sealed chip seal on epoxy asphalt pavement for steel bridge deck

In order to improve the surface performace of epoxy asphalt pavement （EAP） for steel bridge deck, an epoxy asphalt chip seal （ ECS） covered by a cationic emulsified asphalt fog seal （i. e., fog-sealed chip seal） isproposed and a laboratory study is conducted to design and evaluate te fog-sealed chip seal. First, the evaluation indices and methods of te chip seal on steel bridge deck pavement were proposed. Secondly, the worst pavement conditions during te maintenance time were simulated by te small traffic load simulation system MMLS3 and the short-term aging test for minimizing the failure probability of chip seal. Finally, the design parameters of fog-sealed chip seal were determined by the experimental analysis and the performance of the designed fog-sealed chip seal was evaluated in thelaboratory. Results indicate that the proposed simulation method of pavement conditions is effective and the maximal load repetitions on the EAPslab specimen are approximately 925 300 times. Moreover, the designed fog-sealedchip sealcan provide a dense surface with sufficient skid resistance,aggregate-asphalt aahesive performance and interlayer shearing resistance.

Structural Design of Pavement Overlays Based on Functional Parameters

This paper reports a practical pavement overlay design method based on PCI (Pavement Condition Index). Current pavement investigation method (JTJ 073 96) is compared to the ASTM D 5340, which is the standard test method for airport pavement condition evaluation initially developed for US Air Force. The deficiency in the calculation of PCI based on field data in JTJ 073 is discussed. The proposed design method is compared to AASHTO overlay design method with good agreement. The paper concludes with an example illustrating how the existing pavement structural capacity is related to pavement distress survey results. The presented design method can be used in the design for overlay rehabilitation of pavements of highways, urban streets and airports.

Comparison of deep convolutional neural network classifiers and the effect of scale encoding for automated pavement assessment

Deep learning has received a growing interest in recent years for detecting different types of pavement distresses and automating pavement condition assessment.A proper choice of deep learning models is key for successful pavement assessment applications.In this study,we first present a comprehensive experimental comparison of state-of-the-art image classification models to evaluate their performances on 11 pavement objects classification.Our experiments are conducted in different dimensions of comparison,including deep classifier architecture,effects of network depth,and computational costs.Five convolutional neural network(CNN)classifiers widely used in transportation applications,including VGG16,VGG19,ResNet50,DenseNet121,and a generic CNN(as the control model),are tested with a comprehensive pixel-level annotated dataset for 11 different distress and non-distress classes(UCF-PAVE 2017).In addition,we investigate a simple yet effective approach of encoding contextual information with multi-scale input tiles to classify highly random pavement objects in size,shape,intensity,texture,and direction.Our comparison results show that the multi-scale approach significantly improves the classification accuracy for all compared deep classifiers at a negligible extra computational cost.Finally,we provide recommendations of how to improve the classification performance of deep CNNs for automated pavement condition assessment based on the comparison results.

Improvement of the Algorithm for Computing Adjusted Structure Number for Determination of Backlog Maintenance

The pavement strength is very important for the evaluation of backlog maintenance. The current trend in many developing countries used pavement conditions index-PCI in estimating maintenance costs. The PCI can only justify periodic and routine recurrent maintenance. The condition strength is rarely determined in a flexible pavement creating an opportunity for back long maintenance. This paper reports the study conducted to develop and improve the algorithm for estimating the adjusted structure number to estimate the remaining thickness of the flexible pavement. The analysis of eight ways of computing structure numbers from FWD data ware analyzed and found that the improvement of the HDM 3 - 4 models can influence the usefulness of data collected from road asset management in Tanzania. The algorithm for estimating structural numbers from CBR was improved to compute adjusted structural numbers finally used to estimate the remaining life of the flexible pavement. The analysis of the network of about 6900 km including ST and AM was found that 64.72% were very good, 12% were Good, 10% were fair and 7.84% were poor and 5.4% were very poor.

Deep learning based on connected vehicles for icing pavement detection

Slippery road conditions,such as snowy,icy or slushy pavements,are one of the major threats to road safety in winter.The U.S.Department of Transportation(USDOT)spends over 20%of its maintenance budget on pavement maintenance in winter.However,despite extensive research,it remains a challenging task to monitor pavement conditions and detect slippery roadways in real time.Most existing studies have mainly explored indirect estimates based on pavement images and weather forecasts.The emerging connected vehicle(CV)technology offers the opportunity to map slippery road conditions in real time.This study proposes a CV-based slippery detection system that uses vehicles to acquire data and implements deep learning algorithms to predict pavements’slippery conditions.The system classifies pavement conditions into three major categories:dry,snowy and icy.Different pavement conditions reflect different levels of slipperiness:dry surface corresponds to the least slippery condition,and icy surface to the most slippery condition.In practice,more attention should be paid to the detected icy and snowy pavements when driving or implementing pavement maintenance and road operation in winter.The classification algorithm adopted in this study is Long Short-Term Memory(LSTM),which is an artificial Recurrent Neural Network(RNN).The LSTM model is trained with simulated CV data in VISSIM and optimized with a Bayesian algorithm.The system can achieve 100%,99.06%and 98.02%prediction accuracy for dry pavement,snowy pavement and icy pavement,respectively.In addition,it is observed that potential accidents can be reduced by more than 90%if CVs can adjust their driving speed and maintain a greater distance from the leading vehicle after receiving a warning signal.Simulation results indicate that the proposed slippery detection system and the information sharing function based on the CV technology and deep learning algorithm(i.e.,the LSTM network implemented in this study)are expected to deliver real-time detec-tion of slippery pavement conditions,thus significantly eliminating the potential risk of accidents.

Modeling the dynamic performance of transportation infrastructure using panel data model in state-space specifications

In this study,different modeling approaches used in panel data for performance forecast of transportation infrastructure are firstly reviewed,and the panel data models(PDMs)are highlighted for longitudinal data sets.The state-space specification of PDMs are proposed as a framework to formulate dynamic performance models for transportation facilities and panel data sets are used for estimation.The models could simultaneously capture the heterogeneity and update forecast through inspections.PDMs are applied to tackle the cross-section heterogeneity of longitudinal data,and PDMs in state-space forms are used to achieve the goal of updating performance forecast with new coming data.To illustrate the methodology,three classes of dynamic PDMs are presented in four examples to compare with two classes of static PDMs for a group of composite pavement sections in an airport in east China.Estimation results obtained by ordinary least square(OLS)estimator and system generalized method of moments(SGMM)are compared for two dynamic instances.The results show that the average root mean square errors of dynamic specifications are all significantly lower than those of static counterparts as prediction continues over time.There is no significant difference of prediction accuracy between state-space model and curve shifting model over a short time.In addition,SGMM does not obtain higher prediction accuracy than OLS in this case.Finally,it is recommended to specify the inspection intervals as several constants with integer multiples.